{"id":"W2953385524","doi":"10.1101/679779","title":"Differential Treatment Benefit Prediction For Treatment Selection in Depression: A Deep Learning Analysis of STAR*D and CO-MED Data","year":2019,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Treatment of Major Depression","field":"Medicine","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Jewish General Hospital; University of Toronto; Acadia University; McGill University","funders":"","keywords":"Bootstrapping (finance); Population; Artificial intelligence; Depression (economics); Machine learning; Selection (genetic algorithm); Feature selection; Statistics; Medicine; Computer science; Mathematics; Econometrics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005565336,0.0004545631,0.0009690006,0.0007281708,0.0001847299,0.0005547656,0.0006246456,0.0007398902,0.001250358],"category_scores_gemma":[0.01065252,0.0001624008,0.001146401,0.0004876235,0.0004175177,0.0003785208,0.0006866761,0.001147868,0.0001876567],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008413722,"about_ca_system_score_gemma":0.0006747346,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004813838,"about_ca_topic_score_gemma":0.006218478,"domain_scores_codex":[0.9985287,0.000899649,0.00009239354,0.000187272,0.0002077032,0.00008422329],"domain_scores_gemma":[0.9915455,0.006468587,0.0005429848,0.000597854,0.0005701929,0.0002748014],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.009730328,0.002519481,0.3961817,0.0008344241,0.002759529,0.000666805,0.0001644718,0.4046223,0.00412683,0.003116123,0.01965704,0.1556209],"study_design_scores_gemma":[0.0003182328,0.0008309476,0.06025292,0.00004658387,0.0001715076,0.0001890923,0.00004953146,0.9316683,0.002082712,0.002677394,0.001674721,0.00003807518],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.979331,0.0008108105,0.01289044,0.001460233,0.00005201752,0.0001143254,0.003953335,0.0001971037,0.001190646],"genre_scores_gemma":[0.9906659,0.0000814003,0.004946434,0.0002320514,0.00001905973,0.00004892565,0.003514917,0.00001206522,0.0004791816],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005565336,"threshold_uncertainty_score":0.02943265,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0272819785475747,"score_gpt":0.274791904165503,"score_spread":0.2475099256179283,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}